What is the effect of the low fill factor on the photocurrent curve?

If the product of the electric field and the lifetime/mobility were much larger than the active layer thickness, the extraction of photogenerated carriers would depend less on the field; the photocurrent I-V curve would then behave closer to the ideal horizontal response in the region between 0 V and Voc, which would correspond to a higher fill factor.

What is the effect of zero-applied bias?

Under zero-applied bias, the built-in field due to the contact work function offset is insufficient to lead to an appreciable charge transfer between the oleic acid-capped NC and polymer.

What is the power conversion efficiency of a capped NC?

The power conversion efficiency maximum electrical output power/incident light power is about 0.001% at an incident power of 16 mW and decreases with increased power.

Why is the photocurrent faster in the annealed samples?

The faster time response of the photocurrent in the annealed samples Fig. 4 is also likely due to the combined effects of the more efficient charge separation and the improved electron transport properties that result after annealing.

What is the way to improve photovoltaic performance?

Further improvement in the photovoltaic performance of films made with octylamine-capped NCs occurs upon thermally annealing the composite layer at 220 °C for 1 h.

(Open Access) Enhanced infrared photovoltaic efficiency in PbS nanocrystal/semiconducting polymer composites: 600-fold increase in maximum power output via control of the ligand barrier (2005) | S. Zhang

Q: Why is the fill factor low in their devices?

The low fill factor in their devices could be due to either a high series resistance, possibly due to low carrier mobilities, or a low shunt resistance, possibly from nanoscopic pinholes penentrating the films.

Q: What is the role of ligands in photovoltaics?

In photovoltaic devices, rapid and efficient charge separation is needed for subsequent separate transport and extraction of electrons and holes.

Q: What is the effect of annealing on the photoluminescence?

In addition, the authors propose that removal of some ligands from the NC surface occurs during annealing, causing: a reduced spatial separation of the NC and polymer at the heterojunction interface and thus improved charge separation, and b further reduction of the interparticle spacing in the NC phase to improve electron conduction.

Q: How much light was observed upon annealing?

A 140 times increase in dark current was observed upon annealing at 220 °C, indicating improved charge transport after annealing.

Enhanced infrared photovoltaic efficiency in PbS nanocrystal/semiconducting polymer

composites: 600-fold increase in maximum power output via control of the ligand

barrier

S. Zhang, P. W. Cyr, S. A. McDonald, G. Konstantatos, and E. H. Sargent

Citation: Applied Physics Letters 87, 233101 (2005); doi: 10.1063/1.2137895

View online: http://dx.doi.org/10.1063/1.2137895

View Table of Contents: http://scitation.aip.org/content/aip/journal/apl/87/23?ver=pdfcov

Published by the AIP Publishing

Articles you may be interested in

Carrier multiplication in a PbSe nanocrystal and P3HT/PCBM tandem cell

Appl. Phys. Lett. 92, 191107 (2008); 10.1063/1.2920477

Enhanced photovoltaic performance in nanoimprinted pentacene-PbS nanocrystal hybrid device

Appl. Phys. Lett. 92, 093308 (2008); 10.1063/1.2890848

A PbS nanocrystal- C 60 photovoltaic device for infrared light harvesting

Appl. Phys. Lett. 91, 133506 (2007); 10.1063/1.2790730

Carrier transport in PbS nanocrystal conducting polymer composites

Appl. Phys. Lett. 87, 253109 (2005); 10.1063/1.2140885

Photoconductivity from PbS-nanocrystal∕semiconducting polymer composites for solution-processible, quantum-

size tunableinfrared photodetectors

Appl. Phys. Lett. 85, 2089 (2004); 10.1063/1.1792380

This article is copyrighted as indicated in the article. Reuse of AIP content is subject to the terms at: http://scitation.aip.org/termsconditions. Downloaded to IP:

142.150.235.171 On: Tue, 03 Nov 2015 15:42:04

Enhanced infrared photovoltaic efﬁciency in PbS nanocrystal/

semiconducting polymer composites: 600-fold increase in maximum power

output via control of the ligand barrier

S. Zhang, P. W. Cyr, S. A. McDonald, G. Konstantatos, and E. H. Sargent

a兲

Department of Electrical & Computer Engineering, University of Toronto, Toronto, Ontario

M5S 3G4, Canada

共Received 12 April 2005; accepted 4 October 2005; published online 28 November 2005兲

We report a comparison of photoconductive performance of PbS nanocrystal/polymer composite

devices containing either oleic acid-capped or octylamine capped nanocrystals 共NCs兲. The

octylamine-capped NCs allow over two orders of magnitude more photocurrent under −1 V bias;

they also show an infrared photovoltaic response, while devices using oleic acid-capped NCs do not.

Further improvement in the photovoltaic performance of ﬁlms made with octylamine-capped NCs

occurs upon thermally annealing the composite layer at 220 °C for 1 h. The procedure leads to a

200-fold increase in short circuit current, a 600-fold increase in maximum power output, and an

关DOI: 10.1063/1.2137895兴

Composites of semiconductor nanocrystals 共NCs兲 and

conjugated polymers offer promise for fabrication of large-

area optoelectronic devices on ﬂexible substrates using low-

cost solution processing. Control of the organic-inorganic in-

terfaces on the nanoscale is of critical importance in such

systems. In photovoltaic devices, rapid and efﬁcient charge

separation is needed for subsequent separate transport and

extraction of electrons and holes. Organic ligands passivating

the surfaces of NCs endow solubility, yet these ligands are

typically insulating and thus impede charge transfer between

the NC and polymer.

While moderate success has been

achieved in conjugated polymer/NC composite-based solar

cells active in the visible region,

2–5

nearly one-half of the

solar energy reaching the Earth’s surface lies in the infrared

共IR兲 region beyond 700 nm, and it is therefore of great inter-

est to develop IR-sensitive devices.

Recently, IR photovol-

taic devices based on solution-processible nanocomposites

of PbS NCs in poly关2-methoxy-5-共2

⬘

-ethylhexyloxy-

p-phenylenevinylene兲兴共MEH-PPV兲 were reported.

While

these initial results were promising, the devices exhibited

very low efﬁciencies, meriting further optimization.

Reports investigating the effects of annealing on

polymer-based composite photovoltaics typically cite

changes in the morphology of the separate phases as the

cause for improved charge separation or charge mobility.

8–11

In solar cells consisting of pyridine-capped CdSe in

poly共3-hexylthiophene兲, an increase in the external quantum

efﬁciency by a factor of 1.3 to 6 by heating the ﬁlms is

reported.

The inﬂuence of annealing on photovoltaics made

from polymer/NC composites for use in the IR region has

not been explored.

We demonstrate control of the ligand barrier in MEH-

PPV/PbS NC composites by exchanging the oleic acid

ligands for octylamine ligands in solution prior to casting the

ﬁlms, and observe a large improvement in photoconductive

performance. We further demonstrate control of this interface

in the solid state via thermal annealing of the ﬁlms, which

results in up to a 200-fold improvement in short-circuit cur-

rent and 600-fold increase in maximum power output, as

well as a more rapid photoconductive response.

Oleic acid-capped PbS nanocrystals were synthesized

using the method of Ref. 13. Octylamine-capped PbS NCs

were prepared by ligand exhange as previously described.

The exchange procedure leads to a blueshift of the ﬁrst ex-

citonic absorption peak of the NCs due to a ligand etching

phenomenon. Thus, oleic acid-capped NCs with a ﬁrst-

excitonic peak at 1140 nm afforded octylamine-capped NCs

with a ﬁrst-excitonic peak at 985 nm after the exchange pro-

cess. A second set of octylamine-capped NCs was also used

with a ﬁrst-excitonic absorption peak at 1300 nm. The PbS

NC/MEH-PPV blend layers 共80 wt % NCs兲 in the devices

were prepared via spincoating of chloroform solutions onto

indium tin oxide-coated glass. Annealing of ﬁlms was per-

formed on a hotplate for1hinaN

-ﬁlled glove box with

⬍1 ppm residual O

and H

O. Upper contacts 共3mm

兲 were

then deposited by thermal evaporation to form a metal stack

of 30 nm Mg/100 nm Ag/5 nm Au. The dark and photocur-

rents were measured using an Agilent 4155C Semiconductor

Parameter Analyzer. Optical excitation was provided by a

975 nm semiconductor laser operating in continuous wave

mode, with the beam enlarged to a diameter of ⬃3mmbya

lens and focused on the active device area. The response

time of the devices was obtaining by measuring the voltage

under zero-external applied bias across a load resistor placed

in series with the device. The resistance of the series load

resistor was three orders of magnitude smaller than that of

the device under illumination. Thermal gravimetric analysis

共TGA兲 was performed under N

using a TA Instruments SDT

Q600 with a heating rate of 10 ° C min

−1

. Photolumines-

cence spectra were obtained using the system described in

Ref. 14.

Figure 1 plots the current-voltage 共I-V兲 characteristic,

with and without infrared illumination, for devices using PbS

NCs as-synthesized 共oleic acid ligands, ﬁrst-exciton peak at

1140 nm兲 and following exchange 共octylamine ligands, ﬁrst-

exciton peak at 985 nm兲. The data have been normalized to

account for differences in absorption at the excitation wave-

a兲

Author to whom correspondence should be addressed; electronic mail:

ted.sargent@utoronto.ca

APPLIED PHYSICS LETTERS 87, 233101 共2005兲

This article is copyrighted as indicated in the article. Reuse of AIP content is subject to the terms at: http://scitation.aip.org/termsconditions. Downloaded to IP:

142.150.235.171 On: Tue, 03 Nov 2015 15:42:04

length of the NC samples. No thermal annealing was per-

formed on these samples. No detectable photovoltaic re-

sponse is observed in the devices using oleic acid-capped

NCs. In comparison, the device prepared with octylamine-

capped NCs showed 160 times more photocurrent at −1 V, a

short-circuit current 共I

兲 of 250 nA and an open-circuit po-

tential 共V

兲 of 0.47 V.

The longer oleic acid ligands 共comprising an 18 carbon

chain兲 are ⬃2 nm in length, and provide a signiﬁcantly

greater barrier to dissociation of the exciton created in the

nanocrystal at the NC/polymer interface than the much

shorter octylamine ligands 共comprising an 8 carbon chain兲,

which are ⬃1 nm in length. Under zero-applied bias, the

built-in ﬁeld due to the contact work function offset is insuf-

ﬁcient to lead to an appreciable charge transfer between the

oleic acid-capped NC and polymer. Under applied bias, only

a weak photocurrent is observed in this device. In contrast,

charge separation is greatly enhanced in devices based on

NCs capped using the shorter octylamine ligands, leading to

signiﬁcantly greater photocurrent generation. This could be

due to either more efﬁcient tunneling through the ligand bar-

rier, or direct transfer to polymer at bare sites on the NC

surface present after the ligand exchange process. Consistent

with this, a reduction in interparticle spacing after the ex-

change process has been conﬁrmed in transmission electron

microscopy experiments on ﬁlms prepared from pre- and

postexchanged NCs. In addition, the closer interparticle

separation in the octylamine-capped NCs is expected to pro-

vide better electron conduction through the NC phase, and

thus higher charge collection efﬁciency.

In an effort to improve the photovoltaic performance, we

examined the effects of thermal annealing. The results are

shown for a representative set of devices, using octylamine-

capped NCs with a ﬁrst-excitonic absorption peak of 1300

nm, in Fig. 2. Annealing at temperatures above ⬃160 °C

leads to notable changes in I

and photocurrent under bias.

Compared with the unannealed sample, the sample annealed

at 220 °C shows an I

200 times higher, and a product of

•V

共under 400 mW illumination兲 about 600 times higher.

The short-circuit internal quantum efﬁciency of the annealed

samples is about 0.15%, compared to 0.0064% for the best

samples previously reported.

The dependence of the short-circuit current 共 I

兲, open-

circuit voltage 共V

兲, and the ﬁll factor 关共V•I兲

max

/共I

兲兴 on

the incident light power for a device made from a ﬁlm an-

nealed at 220 °C is shown in Fig. 3. The magnitude of the

observed I

and V

depends on the incident light intensity.

Below 150 mW illumination intensity, I

increases linearly

with power, while at higher intensity I

depends sublinearly

on the light power, which may be ascribed to bimolecular

recombination.

The power conversion efﬁciency 共maximum

electrical output power/incident light power兲 is about

0.001% at an incident power of 16 mW and decreases with

increased power. The low absorbance 共⬍0.06兲 at the illumi-

nation wavelength and the low ﬁll factors are causes of the

low-power conversion efﬁciencies. The low ﬁll factor in our

devices could be due to either a high series resistance, pos-

sibly due to low carrier mobilities, or a low shunt resistance,

possibly from nanoscopic pinholes penentrating the ﬁlms.

Furthermore, the devices likely have a low carrier lifetime/

mobility product leading to a strong dependence of the pho-

tocurrent on the electric ﬁeld between 0 V and V

.Ifthe

product of the electric ﬁeld and the lifetime/mobility were

much larger than the active layer thickness, the extraction of

photogenerated carriers would depend less on the ﬁeld; the

photocurrent I-V curve would then behave closer to the ideal

horizontal response in the region between 0 V and V

, which

would correspond to a higher ﬁll factor.

FIG. 1. 共Color online兲 Dark and illuminated 共200 mW兲 I-V curves of de-

vices made with oleic acid-capped 共squares and circles兲 and octylamine

capped 共triangles兲 PbS nanocrystals. Inset shows the same data in a semilog

plot.

FIG. 2. 共Color online兲 Photocurrent I-V curves under 200 mW illumination

for an unannealed sample 共squares兲, and samples annealed at 160 °C

共circles兲, 190 °C 共triangles兲, and 220 °C 共stars兲. Inset shows same data in

semilog plot.

FIG. 3. 共Color online兲 Dependence of the short-circuit current 共I

兲, open-

circuit voltage 共V

兲, and ﬁll factor 关FF= maximum power output/

共I

ⴱ V

兲兴 on the incident optical power for a device annealed at 220 °C.

233101-2 Zhang et al. Appl. Phys. Lett. 87, 233101 共2005兲

This article is copyrighted as indicated in the article. Reuse of AIP content is subject to the terms at: http://scitation.aip.org/termsconditions. Downloaded to IP:

142.150.235.171 On: Tue, 03 Nov 2015 15:42:04

The temporal behavior of I

for the as-deposited and

annealed sample at 220 °C is shown in Fig. 4. Switching the

laser off/on or on/off causes an I

decay to a stable value

following a quick rise, or an I

rise following a quick drop,

respectively, for both samples. The annealed sample reaches

the stable state much more quickly. The I

rise time for the

unannealed sample is 18 ms, compared to 1.7 ms for the

annealed sample. The temporal behavior at zero bias results

from charge trapping and release. In the illuminated state,

more electrons are trapped at the cathode and more holes at

the anode, which screens out the built-in ﬁeld and diminishes

the current. In the dark state, the trapped charges are released

and may move in the opposite direction to return to the equi-

librium state, as shown by the current spike preceding the

settling to the off state.

The improved device performance after annealing is pro-

posed to be due to improved charge separation and improved

charge transport. Both the dark and photocurrents were

found to increase exponentially with annealing temperature,

although the dark current increases at a much lower rate. A

140 times increase in dark current was observed upon an-

nealing at 220 °C, indicating improved charge transport after

annealing. Annealing near or above the glass transition tem-

perature 共⬃215 ° C兲 of the polymer could also allow mor-

phology changes in the ﬁlm that might yield more favorable

electrical transport properties in both polymer and NC

phases. Such morphology changes could thus yield a more

efﬁcient bulk heterojunction structure in the ﬁlm, improving

both charge separation and transport. In addition, we propose

that removal of some ligands from the NC surface occurs

during annealing, causing: 共a兲 reduced spatial separation of

the NC and polymer at the heterojunction interface and thus

improved charge separation, and 共b兲 further reduction of the

interparticle spacing in the NC phase to improve electron

conduction.

We investigated whether heating of the ﬁlms could result

in removal of the ligands from the NC surface. TGA data for

an octylamine-capped sample of PbS NCs shows 5% weight

loss after heating to 200 °C. In comparison, the oleic acid-

capped NCs demonstrated no appreciable weight loss below

300 °C. The TGA data suggest that a certain amount of oc-

tylamine ligand is removed from the ﬁlm during the anneal-

ing process at or above 190 °C 共the boiling point of octy-

lamine is 175 °C兲, permitting closer contact between the

MEH-PPV backbone chains and the nanocrystal surface, and

thus improving the efﬁciency of the charge transfer at the

interface. Complete quenching of the NC photoluminescence

after annealing at 220 °C was also observed, suggesting

rapid exciton dissociation before recombination, and sup-

porting the conclusion of efﬁcient charge separation between

the NC and polymer. The faster time response of the photo-

current in the annealed samples 共Fig. 4兲 is also likely due to

the combined effects of the more efﬁcient charge separation

and the improved electron transport properties that result af-

ter annealing. Further improvement to the efﬁciency of these

devices employing the thermal annealing process may be

possible using different ligands, altering the annealing con-

ditions, and selecting conjugated polymers with more favor-

able hole accepting and/or hole mobility properties to reduce

the recombination of charge carriers in the device.

The authors gratefully acknowledge the support of Ma-

terials and Manufacturing Ontario, a division of the Ontario

Centres of Excellence; the Natural Sciences and Engineering

Research Council of Canada under the Collaborative Re-

search and Development Program; Nortel Networks; the

Canada Foundation for Innovation; the Ontario Innovation

Trust; and the Canada Research Chairs Programme, as well

as the Government of Ontario through the Ontario Graduate

Scholarships program for one of the authors 共S.A.M.兲. The

authors would like to thank Dr. L. Levina for synthesis of

nanocrystals and L. Vanderark, Dept. of Chemistry, Univer-

sity of Toronto for assistance with TGA.

N. C. Greenham, X. Peng, and A. P. Alivisatos, Phys. Rev. B 54, 17628

共1996兲.

W. U. Huynh, X. Peng, and A. P. Alivisatos, Adv. Mater. 共Weinheim, Ger.兲

11, 923 共1999兲.

W. U. Huynh, J. J. Dittmer, and A. P. Alivisatos, Science 295, 2425

共2002兲.

B. Sun, E. E. Marx, and N. C. Greenham, Nano Lett. 3,961共2003兲.

W. J. E. Beek, M. M. Wienk, and R. A. J. Janssen, Adv. Mater. 共Weinheim,

Ger.兲 16, 1009 共2004兲.

E. H. Sargent, Adv. Mater. 共Weinheim, Ger.兲 17,515共2005兲.

S. A. McDonald, G. Konstantatos, S. Zhang, P. W. Cyr, E. J. D. Klem, L.

Levina, and E. H. Sargent, Nat. Mater. 4,138共2005兲.

N. Camaioni, G. Ridolﬁ, G. Casalbore-Miceli, G. Possamai, and M.

Maggini, Adv. Mater. 共Weinheim, Ger.兲 14, 1735 共2002兲.

J. J. Dittmer, R. Lazzaroni, P. Leclère, P. Moretti, M. Granström, K.

Petritsch, E. Marseglia, R. H. Friend, J. L. Brédas, H. Rost, and A. B.

Holmes, Sol. Energy Mater. Sol. Cells 61,53共2000兲.

J. J. Dittmer, E. A. Marseglia, and R. H. Friend, Adv. Mater. 共Weinheim,

Ger.兲 12, 1270 共2000兲.

W. Feng, A. Fujii, S. Lee, H. Wu, and K. Yoshino, J. Appl. Phys. 88, 7120

共2000兲.

W. U. Huynh, J. J. Dittmer, W. C. Libby, G. L. Whiting, and A. P.

Alivisatos, Adv. Mater. 共Weinheim, Ger.兲 13,73共2003兲.

M. A. Hines and G. D. Scholes, Adv. Mater. 共Weinheim, Ger.兲 15,1844

共2003兲.

T.-W. F. Chang, A. Maria, P. W. Cyr, V. Sukhovatkin, L. Levina, and E. H.

Sargent, Synth. Met. 148, 257 共2005兲.

FIG. 4. Temporal behavior of I

for an unannealed sample 共bottom兲 and a

sample annealed at 220 °C 共top兲.

233101-3 Zhang et al. Appl. Phys. Lett. 87, 233101 共2005兲

This article is copyrighted as indicated in the article. Reuse of AIP content is subject to the terms at: http://scitation.aip.org/termsconditions. Downloaded to IP:

142.150.235.171 On: Tue, 03 Nov 2015 15:42:04

Enhanced infrared photovoltaic efficiency in PbS nanocrystal/semiconducting polymer composites: 600-fold increase in maximum power output via control of the ligand barrier

Figures

Citations

Prospects of Colloidal Nanocrystals for Electronic and Optoelectronic Applications

Mesoscopic CH3NH3PbI3/TiO2 heterojunction solar cells.

Schottky Solar Cells Based on Colloidal Nanocrystal Films

Vertically-aligned nanostructures of ZnO for excitonic solar cells: a review

Efficient, stable infrared photovoltaics based on solution-cast colloidal quantum dots.

References

Hybrid Nanorod-Polymer Solar Cells

Solution-processed PbS quantum dot infrared photodetectors and photovoltaics

Colloidal PbS Nanocrystals with Size-Tunable Near-Infrared Emission: Observation of Post-Synthesis Self-Narrowing of the Particle Size Distribution

Charge separation and transport in conjugated-polymer/semiconductor-nanocrystal composites studied by photoluminescence quenching and photoconductivity.

Efficient hybrid solar cells from zinc oxide nanoparticles and a conjugated polymer

Related Papers (5)

Hybrid Nanorod-Polymer Solar Cells

Solution-processed PbS quantum dot infrared photodetectors and photovoltaics

Charge separation and transport in conjugated-polymer/semiconductor-nanocrystal composites studied by photoluminescence quenching and photoconductivity.

Air-Stable All-Inorganic Nanocrystal Solar Cells Processed from Solution

Colloidal PbS Nanocrystals with Size-Tunable Near-Infrared Emission: Observation of Post-Synthesis Self-Narrowing of the Particle Size Distribution

Frequently Asked Questions (21)

Q1. Why is the fill factor low in their devices?

Q2. What is the role of ligands in photovoltaics?

Q3. What is the effect of annealing on the photoluminescence?

Q4. What is the effect of the light on the electrons?

Q5. What is the effect of the low fill factor on the photocurrent curve?

Q6. What is the effect of zero-applied bias?

Q7. What is the role of ligands in the photovoltaics?

Q8. How was the response time of the devices measured?

Q9. How much light was observed upon annealing?

Q10. What is the power conversion efficiency of a capped NC?

Q11. What is the reason for the low fill factor in the devices?

Q12. What is the effect of annealing on photovoltaics?

Q13. What is the optimum photovoltaic performance of a PbS nanocry?

Q14. What is the effect of annealing on the efficiency of the NC?

Q15. How is the short-circuit internal efficiency of the annealed samples?

Q16. Why is the photocurrent faster in the annealed samples?

Q17. Why is the charge separation significantly enhanced in NCs capped?

Q18. What is the way to improve photovoltaic performance?

Q19. Why is the low fill factor in the annealed sample better?

Q20. What is the reason for the low fill factor in the annealed sample?

Q21. How does the annealing of the films affect the photoconductive response?